fs/file.c: don't acquire files->file_lock in fd_install()

Mateusz Guzik reported :

 Currently obtaining a new file descriptor results in locking fdtable
 twice - once in order to reserve a slot and second time to fill it.

Holding the spinlock in __fd_install() is needed in case a resize is
done, or to prevent a resize.

Mateusz provided an RFC patch and a micro benchmark :
  http://people.redhat.com/~mguzik/pipebench.c

A resize is an unlikely operation in a process lifetime,
as table size is at least doubled at every resize.

We can use RCU instead of the spinlock.

__fd_install() must wait if a resize is in progress.

The resize must block new __fd_install() callers from starting,
and wait that ongoing install are finished (synchronize_sched())

resize should be attempted by a single thread to not waste resources.

rcu_sched variant is used, as __fd_install() and expand_fdtable() run
from process context.

It gives us a ~30% speedup using pipebench on a dual Intel(R) Xeon(R)
CPU E5-2696 v2 @ 2.50GHz

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Mateusz Guzik <mguzik@redhat.com>
Acked-by: Mateusz Guzik <mguzik@redhat.com>
Tested-by: Mateusz Guzik <mguzik@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

1 files changed, 4 insertions, 0 deletions

diff --git a/Documentation/filesystems/porting b/Documentation/filesystems/porting
index 3eae250254d5..ec5456113072 100644
--- a/Documentation/filesystems/porting
+++ b/Documentation/filesystems/porting
@@ -500,3 +500,7 @@ in your dentry operations instead.
         dentry,  it does not get nameidata at all and it gets called only when cookie
         is non-NULL.  Note that link body isn't available anymore, so if you need it,
         store it as cookie.
+--
+[mandatory]
+        __fd_install() & fd_install() can now sleep. Callers should not
+        hold a spinlock or other resources that do not allow a schedule.